What exactly IS the overpopulation argument (in regards to immortality)?

by Raemon1 min read13th Mar 201155 comments

11

Personal Blog

Periodically in discussions of cryonics and related issues, people bring up "the overpopulation argument, and the counterarguments that respond to it" without actually describing those arguments in detail. Overpopulation was a big concern of mine prior to exposure to anti-deathism. And so far, it still is. I have no philosophical problem with eliminating death, but it seems to me that getting rid of death BEFORE we eliminate all the other major problems facing humanity is going to make those problems much worse. Death is an enemy that should be vanquished, but it's not an enemy I'm prepared to destroy until I'm satisfied that we'll be able to handle the consequences.

If death is solved via uploads running at high speed, I'm not too concerned. (Still a little concerned, since computers still take up space, but the issue is close enough to negligible that I'm fine ignoring it).

If we're dealing with physical humans taking up physical space requiring physical resources, then I'm worried. Either we're growing exponentially, or we've eliminated childbirth, or we have strict rules in place about people who have children being willing to die. (The latter might work but modifying central aspects of the human life cycle that we are hard-wired to value seems.... challenging, to say the least)

The sense of I've gotten around here is that "exponential growth is okay, because Space is Big". Space certainly is big, and I imagine we could expand for a long time without running into conflict. But if there's even one other alien race who solves their problems the same way (expanding whenever they run out of space or resources), then eventually there's going to be conflict. And the longer we go BEFORE that conflict, the more human suffering it might entail. I'd prefer to have achieved equilibrium as a species beforehand. 

I want to expand into the universe, but I think we should do so out of *curiosity* rather than *necessity.* 

I assume there's been a lot of discussions about this and I don't want to rehash them. But if someone could summarize the issues at hand and explain what the community consensus is, I'd appreciate it. ("Community Consensus" might mean "there are a few dominant schools of thought here").

Edit: FAWS pointed out that as long as, on average, each person reproduces slightly less than once, growth will not continue exponentially. That's pretty much the answer I'm looking for. The logistics are still significant, but in the long run I think such a law would be enforcible. (Each person only gets to reproduce once, and some people will choose not to. I think there'd need to be an additional disincentive, because over the course of an immortal lifespan, people are likely to try out childrearing at *some* point).

I still think that the issues surrounding overpopulation should occupy at least as much of our attention as ending death (basically ensuring that people are given adequate resources to live healthy, productive lives).

11

55 comments, sorted by Highlighting new comments since Today at 9:28 AM
New Comment

With no immortality, population growth is exponential. With everyone immortal, population growth is a slightly faster exponential. It's a Problem that has to be solved no matter what, in either case, or else things get Bad. In both cases, the minimum sufficient measure is artificially capping reproduction. With immortality, the cap (as measured in children per parent) is somewhat lower.

Exponential growth is okay for awhile, but there will always come a point when it's not. There are hard limits imposed on our expansion by the speed of light, and they are only cubic, not exponential.

here are hard limits imposed on our expansion by the speed of light, and they are only cubic, not exponential.

Not if space is hyperbolic. The volume of a hyperbolic sphere is exponential in its radius.

Actually, even if space is Euclidean, the effects of special relativity still mean that we can expand exponentially (for as long as we're able to maintain some constant amount of acceleration in any direction we choose, that is). In Minkowski spacetime, the subset of the future light cone consisting of points where a fixed amount of proper time has elapsed is in fact a hyperbolic 3-space.

(But for all practical purposes, you're right about the unsustainability of exponential growth.)

Actually, even if space is Euclidean, the effects of special relativity still mean that we can expand exponentially (for as long as we're able to maintain some constant amount of acceleration in any direction we choose, that is).

I'm curious what would be the requirements in terms of energy and mass to maintain this motion under special relativity. In Newtonian physics, constant acceleration of a constant mass requires power increasing linearly with time -- are things substantially different under SR? Also, even if free energy is somehow given, is it possible to maintain constant acceleration without constantly losing mass in the other direction due to momentum conservation (assuming you're not moving along infinite-mass rails of some sort)?

are things substantially different under SR?

Well, the amount of energy needed seems to be linear in the Lorentz factor, which is rate of change of co-ordinate time with respect to proper time. But the latter is increasing exponentially, so I think you need power increasing exponentially with time to maintain constant acceleration.

Also, even if free energy is somehow given, is it possible to maintain constant acceleration without constantly losing mass in the other direction due to momentum conservation (assuming you're not moving along infinite-mass rails of some sort)?

No - you'd need to somehow keep on propelling yourself. I suppose in theory you could harvest forward momentum from any stars and galaxies you meet, no matter how rapidly you're whizzing past them. That's what's going to need exponentially increasing power.

(Somehow I don't think this is going to work!)

A lot of this depends on the subtleties of the laws of physics. I think the best solution is to end death now; if the math doesn't work out in our favour, we can always just slow down subjective time and build our own exponentials.

A lot of this depends on the subtleties of the laws of physics.

Oh sure. In any case, the "exponential growth in a Euclidean space" scenario wouldn't work because we wouldn't be able to maintain constant forward acceleration indefinitely, if everything around us was moving backwards at close to light speed.

I think the best solution is to end death now; if the math doesn't work out in our favour, we can always just slow down subjective time and build our own exponentials.

The last bit may not be possible. There's a finite limit to the amount of information one can store in a given region of space (or else the amount of mass-energy needed to represent it would collapse into a black hole). So to have exponentially many people, you'll always need an exponential amount of space, whatever you do to subjective time.

If we slow down subjective time at an always increasing rate, we could ensure that the amount of people that we can support increases at the same rate as the amount of people we need to support.

I see. That does the keep space requirements down, but subjectively we'd be moving along this graphical timeline at constant speed. I don't how much computation is possible in the 'Dark Era'. (Perhaps only a finite amount?)

I don't know if that's more or less of a problem than death or limiting births. That is a question for a FAI; while we may be able to solve it, only an AI would need to. I doubt that the concept of personal identity will even survive the singularity, so this could easily end up not mattering.

[-][anonymous]10y 0

Oh sure. I regard the above purely as a mathematical curiosity.

It doesn't provide exponential growth for very long, because the people at the frontier of this expanding hyperbolic sphere would reach the heat death of the universe alarmingly soon. At least, I think that's what would happen, assuming the universe will end in a heat death (as opposed to a Big Rip or Big Crunch).

ETA: Actually, that only defeats the 'exponential growth in a Euclidean space' scenario. But still, I agree that we can't blithely assume current physics.

This is a big complex issue, which in my opinion mostly boils down to the fact that the hypothesis that immortality will cause overpopulation shouldn't be privileged over the hypothesis that it will lead to better population control. Some points to consider:

  • Wanting to have kids (to the extent of being willing to cause environmental catastrophe by so doing) is plausibly in part related to the unfulfilled desire to live longer. Reproduction allows one to have an impact on human society which is meaningful over long time scales. Indefinite life extension fulfills this need without increasing the number of individuals.
  • Having more than one surviving child per adult on average causes exponential growth problems even if immortality goes unsolved. The solution to this is birth control. The automated natural death cycle, aside from its inhumane aspects, is also arguably a fake solution that deludes people into thinking that contraception and conscious family planning is unnecessary.
  • Humans die of accidents and suicide every so often to begin with. Thus there will be new space made for a small number of families to have new children. It will simply not be a major part of the overall culture.
  • Space is big enough to mean there is a lot of time to solve the problem if we limit ourselves to natural biological reproduction. Biotech that allows gigantic family sizes with little cost to the parents (e.g. food factories and artificial wombs) is not the result of immortality but of general advancements in biotechnology.
  • An engineered lifecycle wherein an individual returns to childhood after a period of elderhood is a plausible substitute for the existing status quo of all children being newly created as blank slates. Children could thus exist, but with a rich set of memories and experiences to draw upon and integrate into their lives.
  • The desire to reproduce is largely a cultural artifact, as opposed to an innate drive. To the extent that it is an innate drive, it may be fulfilled by raising any kind of infant-like creature -- including reborn adults, simulated children, and animals.
  • Sometimes it is acceptable to focus on near-term problems at the expense of the extremely long-term. All forms of life-saving human technology arguably contribute to human overpopulation, but given the severity of death it is a worthwhile expenditure. We would rather have a large population than have millions die of hunger or disease -- why should aging be different?
  • Individuals moving to a silicon substrate would most likely not consume anywhere near the resources that natural humans do.
  • Individuals could also choose to move their brain and central nervous system to a low energy-cost nourishment tank, and use robotic bodies that cost far less energy than a natural human body.
  • Groups could plausibly merge into single individuals, by forming hive-minds, effectively reducing the total population.

It may be relevant that on reading some of your more science-fictiony proposed replacements, my thoughts were along the lines of "geez, I hope these take long enough to implement that I get to be a real mother before anyone expects me to find these facsimiles satisfactory".

To be honest I feel the same way. But we're talking about a fairly long period of cultural (not to mention biological) evolution before this becomes a problem, which is part of my point. The people of thousands of years in the future may well consider things that we today find almost too weird to contemplate, to be simply common sense.

Wanting to have kids (to the extent of being willing to cause environmental catastrophe by so doing) is plausibly in part related to the unfulfilled desire to live longer. Reproduction allows one to have an impact on human society which is meaningful over long time scales. Indefinite life extension fulfills this need without increasing the number of individuals.

Wanting to eat (to the extent of being willing to cause environmental catastrophe by doing so) is plausibly in part related to the unfulfilled desire to extract free energy from the surroundings in order to remain far from equilibrium with them. Eating high-energy, body-compatible nutrients allows one to effect such an extraction. Receiving nutrients via IV fulfills this need without environmental cost.

Therefore, apes on nutrient feeds no longer wish to eat. (???)

Receiving nutrients via IV fulfills this need without environmental cost.

I wonder how you imagine nutrients being produced without environmental cost? Of course, one would be distant from their production; but here in the West one is usually distant from the production of ordinary foodstuffs also. And I think your argument confuses adapting with executing adaptations.

Therefore, apes on nutrient feeds no longer wish to eat. (???)

The reasoning may be doubtful, but the conclusion is actually true. Anorexia is a recognised, if minor, side-effect of sustaining someone through an IV when they cannot eat. When they come off the IV, it can take a few days for the normal sensations of hunger and appetite to function properly again.

The desire to reproduce is largely a cultural artifact, as opposed to an innate drive. To the extent that it is an innate drive, it may be fulfilled by raising any kind of infant-like creature -- including reborn adults, simulated children, and animals.

I find this highly implausible, from anything resembling an ev-psyc point of view. Note for example that even animals that don't have culture have an innate drive to reproduce.

I find this highly implausible, from anything resembling an ev-psyc point of view. Note for example that even animals that don't have culture have an innate drive to reproduce.

Smart animals generally have a drive to have sex, not a drive to reproduce. Evolution didn't anticipate (because that blind idiot god never anticipates) that some species would be so so smart that they could easily fulfill the sex drive without reproducing. However, at this point, given the ease of birth control in much of the world, there should be direct selection pressure for wanting to reproduce, not just have sex.

As long as some people exist who tend to actually reproduce - whether or not they consciously want it - they will outbreed those who don't reproduce and dominate the population.

Any solution proposing to change people (biologically, culturally, etc) so that they can reproduce but usually don't, must make very sure there are absolutely no heritable exceptions.

JoshuaZ is basically correct- we evolved to pursue sex and to love kids once born, not to pursue reproduction in the abstract. Relevant LW article.

My essay, overpopulation is a non-issue for radical life extension:

One of the most common objections against the prospect of radical life extension (RLE) is that of overpopulation. Suppose everyone got to enjoy from an eternal physical youth, free from age-related decay. No doubt people would want to have children regardless. With far more births than deaths, wouldn't the Earth quickly become overpopulated?

There are at least two possible ways of avoiding this fate. The first is simply having children later. Even if nobody died of aging, there would still be diseases, accidents and murders. People who've looked at the statistics estimate that with no age-related death, people would on average live to be a thousand before meeting their fate in some way. Theoretically, if everyone just waited to be a thousand before having any kids, then population growth would remain on the same level as it is today.

Of course, this is completely unrealistic. Most people aren't going to wait until they are a thousand to have kids. But they might still have them considerably later than they do now. The average age for having your first child has already gone up as lifespans have grown. If you're going to live for a thousand years, why rush with having kids as soon as possible?

Currently there is (at least for women) an effective maximum cap on how high the age for first childbirth can grow, since once a mother's age grows beyond 35 or so, the probability for birth defects goes up radically. However, current reproductive technology has already made pregnancies over the age of 50 a real possibility. At the moment, this frequently requires egg donation, but a rudimentary ability to produce eggs from stem cells may not be that far away, certainly a lot closer than RLE. By the point that we have RLE, we'll likely also have the ability to produce new sperm and eggs from a person's own cells. Combined with an overall better condition of the body brought about by RLE, this seems like it could increase the maximum age for pregnancy indefinitely. With that, the average age for a first birth going up at least a couple of decades doesn't seem all that unrealistic.

Besides the average age for having kids going up, there's the possibility of larger family groups. Must we necessarily have a norm for children being the kids of exactly two adults? As a personal example, my best friend has a daughter who's two years old right now. I've been over there helping take care of the girl a lot, enough to make me feel like she's part of my family as well. Even if I never had children of my own, I already feel something resembling the feelings related to having a child of your own. In addition to growing attached to the children of your close friends, polyamory is also gradually becoming more common and accepted. With romantic relationships involving more than two people we also get children with more than two parent-like figures. Many have a strong desire to pass on their genes, something which can be helped with e.g. the recent creation of 3-parent human embryos.

So with both the prospect of having kids later and a child having more than two parents, I really don't think that the population problem is as hard to solve as some people make it out to be. It should also be noted that it's not like scientists are going to develop RLE one day, and then the next, blam, everyone lives forever. Rather, the technology will be developed in stages. In the early stages, there are going to be a lot of people who have grown far too frail to be helped, and it might take a long time before we hit acturial escape velocity, so there might simply be an e.g. 10-year bump on people's lifespan and then 20 years could pass before the next major breakthrough.

The treatments may also not be affordable for everyone at first, though it needs to be noted that governments will have a huge incentive to subsidize the treatements for everyone to reduce the healthcare costs of the elderly and to push back the age for retirement. A 2006 article in The Scientist argues that simply slowing aging by seven years would produce large enough of an economic benefit to justify the US investing three billion dollars annually to this research. The commonly heard "but only the rich could live forever" argument against RLE does not, I feel, take into account the actual economic realities (amusingly enough, as its supporters no doubt think they're the economically realistic ones).

So we're going to get a slowly and gradually lengthening average lifespan, which at first probably won't do much more than reverse the population decline that will hit a lot of Western countries soon. The replenishment rate required to keep a population stable is about 2.1 children per woman. The average fertility rate in a lot of industrialized countries is well below this - for instance, 1.58 in Canada, 1.42 in Germany, 1.32 in Italy, 1.20 in Japan and 1.04 in Hong Kong. The EU average is 1.51. Yes, in a lot of poor countries the figures are considerably higher - Niger tops the chart with 7.68 children per woman - but even then the overall world population growth is projected to start declining around 2050 or so.

To give a sense of proportion: suppose that tomorrow, we developed literal immortality and made it instantly available for everyone, so that the death rate would drop to zero in a day, with no adjustment to the birth rate. Even if this completely unrealistic scenario were to take place, the overall US population growth would still only be about half of what it was during the height of the 1950s baby boom! Even in such a completely, utterly unrealistic scenario, it would still take around 53 years for the US population to double - assuming no compensating drop in birth rates in that whole time.

We've adapted to increasing lifespans before. Between 1950 and 1990, the percentage of population over 65 almost doubled in Sweden, going from 10.3 to 18.1. (In the United Kingdom it went up from 10.7 to 15.2, in the US from 8.1 to 12.6, and in the more-developed countries overall it went from 7.6 to 12.1.) The beauty of economics is that like all resource consumption, having children is a self-regulating mechanism: if a growing population really does exert a heavy strain on resources, then it will become more expensive to have children, and people will have less of them. The exception is in the less industrialized countries where children are still a net economic benefit for their parents and not a cost, but most of the world is industrializing quickly. Over the last fifty years, the gaps between the rich and poor have gotten smaller and smaller, to the point where people are calling the whole concept of a first world/third world divide a myth. I see no reason to presume that radical life extension and indefinite youths would pose us any problems that we couldn't handle, at least not on the overpopulation front.

For anyone curious, this was mostly an English recap of some of the life extension-related discussion I covered in my 2009 book Kehittyvä ihmiskunta. If you can't read Finnish but were wondering what I wrote in that book, well, now you know a bit.

Raemon:

If death is solved via uploads running at high speed, I'm not too concerned. (Still a little concerned, since computers still take up space, but the issue is close enough to negligible that I'm fine ignoring it).

Have you read Robin Hanson's writings on the economics of uploads? He argues very convincingly that uploads will inevitably lead to a Malthusian equilibrium, in fact much more rapidly than a biological immortality scenario.

The sense of I've gotten around here is that "exponential growth is okay, because Space is Big".

It can't possibly be okay, since sustained exponential growth must eventually catch up with the expansion of our light cone, which is (barring faster than light travel) the fundamental physical limit on how much space can be available. Exponential growth always dominates polynomial growth given enough time, and I'm sure most people here are well aware of that, so you probably got a wrong impression.

The impression was given more from random one-off comments that were clearly optimized for poetry rather than factual/moral accuracy. So it's very possible I got the wrong impression. But I've heard a lot of noise about eliminating death, and basically zero noise about fixing overpopulation.

There's nothing wrong with trying to do both, but we should be working at least as hard at solving overpopulation as we are on increasing lifespan. I realize that most people here are probably not actually going to contribute to either of those fields, but a disproportion fixation on anti-aging seems dangerous to me. Are the cryonics proponents here making an effort to use minimal resources and take up minimal real estate?

Have you read Robin Hanson's writings on the economics of uploads?

I haven't. Can you provide a link?

Hanson's IEEE Spectrum article is a good start. See also this discussion on Overcoming Bias and its followup. There are also numerous other posts in OB archives discussing these issues; some of them are under the tag "ems."

The economist Nick Rowe recently wrote another good analysis along similar lines.

If we're dealing with physical humans taking up physical space requiring physical resources, then I'm worried. Either we're growing exponentially, or we've eliminated childbirth, or we have strict rules in place about people who have children being willing to die. (The latter might work but modifying central aspects of the human life cycle that we are hard-wired to value seems.... challenging, to say the least)

This is not true. As long as the average total number of children per person stays below 2 (and every child has at least 2 parents) population growth will be sub-linear and approach a finite limit. For example if the average is 1.99 population will never exceed 200 times the starting population, and may take quite a long time to reach that level if there is a strong social expectation of waiting with children until you have attained the necessary maturity to become parents, well after your tenth millennium. I don't think limiting people to just two children is excessively draconian.

Can you explain the math here? I'm pretty confused about how a system in which people give birth but don't die could ever have a natural population limit. (If that's not the part that's not true, can you clarify?)

As for social pressures not to give birth until your 10th millenium... how long do you think it'll take for such a norm to become commonplace? (This is an honest question I don't know the answer to, not an attempt at socratic communication)

In the past century or so, we've seen major drops in how often childbirth happens in developed countries, and how long people wait to have children. There seems to be a correlation between good healthcare and women's equality and people choosing to have fewer children. So I have some hope that in a world of immortals, we'd see similar trends and a continuing drop in birthrates. But I also would never have predicted the drop that we've seen so far, and I'm wary of gambling our future on similar predictions.

Can you explain the math here? I'm pretty confused about how a system in which people give birth but don't die could ever have a natural population limit. (If that's not the part that's not true, can you clarify?)

Start with 10 billion people in generation 1. They have 10 billion * 1.99/2= 9.95 billion children, spread out over however long generation 1 takes to exhaust its allocation. Generation 2 ( 9.95 billion people) has ~9.9 billion children, generation 3 ~9.85 billion children, generation 100 ~6 billion children, generation 500 ~0.82 billion children, generation 1000 ~66.5 million children, generation 2000 ~443 thousand children, and generation 4593 only one child.

As for social pressures not to give birth until your 10th millenium... how long do you think it'll take for such a norm to become commonplace?

A few ten thousand years, at least. With child birth being delayed longer and longer in between.

This is pretty much the answer I was looking for. I've updated. Thanks.

Thank you. I was growing increasingly worried as I read this threadthat no one had bothered with math. Each personhaving 1 children with the small subtraction of those whohave none doesn't seem a draconian price for immortality.

Whenever I hear people refer to death as a tool that we can't toss aside until other problems have been fixed I want to kill them, personally, and ask them if that's what they had in mind. Well, alright, that's an exaggeration, but I want to point out that they are thinking of death as some far-away, abstract thing that happens to OTHER PEOPLE that they don't care about.

Scenario: You will die later today. Look at the clock. Before it hits midnight you will perish. Now stop looking at the clock, you're burning precious seconds.

But wait! There's a way out - yes, you already guessed it, you can cure death completely for everyone! Is it now so terribly important that all those Big Problems that you just had to hold on to Death to address be fixed before death is cured? Are you willing to die within hours because we need Death so badly to keep our Big Problems in check? Or are you ok with releasing the cure for death so you can continue breathing and get around to fixing those problems over the next few years/decades?

If you aren't willing to kill yourself today to keep the Big Problems under control then you cannot ask others to die for that reason.

While it is FAR from the worst I can imagine anyone having to go through, my two dogs both just died within a month of each other (I literally dug the second grave a day before posting this thread), and last year my grandmother died. I don't know whether dogs matter in your moral schema. They matter in mine, whether or not they comprehend what exactly is happening.

All three died of old age. Their deaths (all three of them) did not affect me much. What affected me was watching them suffer for the year(s) prior to their death. Watching them lose their ability to control their bodies, suddenly falling down for no reason. Having to rely on other people to help them walk, then relying on other people to feed them, and finally (in my grandmother's case) having to rely on other people to think for them. She lost her memory, she lost her rationality, she lost her ability to even MAKE decisions at all. I could see the humiliation and horror on her face as she lost control of everything that defined her as a person. I gradually stopped thinking of her as a person, because she stopped demonstrating the traits that I associate with people.

One day, towards the end of her life, I was sitting in her hospital room because I felt obligated to. I talked a little bit to her in case she could hear me and in case it gave her comfort. She didn't give any sign that she could hear me. Apparently she was asleep. I said "okay, I'm gonna go find Mom now. I'll be back later."

All she had the energy to do was squeeze my hand and say "no, please stay." It was one of the last expressions of desire I heard her make. And I was horrified that inside her failing body and mind was a person who wanted, if nothing else, to be loved. And terribly ashamed that I had stopped believing that.

Nobody should have to experience that.

Nobody should have to die either. I didn't say the should. I explicitly stated that I just wanted to know WHY people around here didn't seem so concerned about the Big Problems, just because there was a Bigger Problem. And by the time you made this post, FAWS had answered my question and I said so.

But honestly, it wasn't my grandmother's death that moved me to tears at her funeral. I can come up with justifications for that, I dunno if they'd hold up to rational scrutiny. "She had an amazing life and did everything she set out to do" and all that jazz. It just didn't impact me. The thing that made me cry was the horrible, creeping living death she experienced in the last years of her life. Give me a choice between ending suffering with painless deaths in sleep, or ending death, and I will choose to end suffering in a heartbeat. So long as there are new people to experience joy and fulfillment.

SECTION BREAK

(because experience tells me this post is too long to not have a section break)

Oddly enough (THIS IS IN NO WAY AN OPINION ON WHAT THE AVERAGE PERSON CAN OR SHOULD FEEL), at this specific moment in my life, I actually am only marginally afraid of dying. I'm 24. Theoretically I have my whole life ahead of me. But I feel like I just wrapped all the previous plot threads of my life, and whatever I do next is starting an entire new story that I've barely even thought about. The things about death that actually SCARED me was the notion that I might suddenly disappear before I got to finish the things I wanted to do. That's what I think is actually unfair - dying without warning.

Right now I'm in limbo, and if you had a concrete, flawless proof that I could sacrifice myself to achieve a greater good... I dunno. Maybe. Some people actually ARE willing to sacrifice themselves for the greater good when the time comes, and others aren't, and I don't know what kind of person I am in that regard. Put a gun to my head and say "end death now!" and I might. But you could also point a gun to my head and say "I'll pull the trigger unless you agree to torture 10 random people you don't know for 50 years" and... who knows, maybe I'm the sort of person who'd do it? (I think I wouldn't. But I think I WOULD if you put a gun to my sister's head). Fight or flight responses are not the same as moral judgments.

In the meantime, my solution to the dilemma is to be a good enough person that sacrificing myself would be LESS effective at solving the Big Problems.

I'm not done thinking about this. It's a complicated issue that I've only recently become aware of. And in the meantime, please do not assume that these are far away issues that I'm just thinking about abstractly.

If my epic post is too long, the short version:

Scenario: You will suffer from malnutrition, living in painful agony, for eternity. You can end it at any time by introducing death to the world. Do you do it?

If you aren't willing to suffer, don't ask others to. But more importantly:

Scenario: You (or a loved one of yours) will die in the next hour. There's a button you can press, which will delay your death by an hour but cause some random person to suffer horribly for an hour each time you press it. How many times do you press the button?

Fight or Flight responses are not moral judgements.

I read both your posts fully, and I apologize for having misjudged you. In my experience I had only run into people who hadn't had any meaningful experience with death making this sort of "but it's good for the world" claim. I should've known better than to generalize that to Less Wrong posters.

Your point is well taken, thank you.

Thanks to you as well.

it seems to me that getting rid of death BEFORE we eliminate all the other major problems facing humanity is going to make those problems much worse. Death is an enemy that should be vanquished, but it's not an enemy I'm prepared to destroy until I'm satisfied that we'll be able to handle the consequences.

I think perhaps you don't realize what "getting rid of death" means. Humanity grew up in the shadow of death, and lives every day in the shadow of death. The whole world population of 1880, and every earlier year, is completely dead. Getting rid of death means that the end never comes, for anyone, beyond a certain point in time. Rejuvenation doesn't get rid of death, it just means you're still young when you do finally die. Cryonics, uploading, and copying don't get rid of death - in fact copying multiplies death - if you live on through your copies, that means you will die many times. And physics says there will be a last time, even in that scenario.

Also, you don't realize the enormity of the mass psychological inertia here. There is no popular will to "get rid of death". If you make that your goal, you will find yourself in a desert, bereft of support and understanding. Humans can be classified into those who don't yet understand the idea of death, those who are resigned to it, and those who believe in a metaphysical safety net. The people who see death as something which the human race could choose to abolish by an appropriate effort are a microscopic minority. Since the first cryonic suspension, over a billion people died, and less than a thousand were frozen; that tells you how small a minority.

The next tipping point will come when there's popular interest in rejuvenation, that is, in getting rid of "ageing". That is a far more doable thing, than getting rid of death per se, and it has an obvious appeal for anyone who is, in fact, ageing. Again, it's only strange minorities who want to take a one-way liquid-nitrogen ticket to an unknown future, or who think of getting their brains scanned so they can live on in cyberspace. But simply staying young would appeal to a lot of people - I won't say everybody, or even a majority, because apparently you can't underestimate the human will to find its existing situation satisfactory and not in need of change.

But returning to your desire to solve "all the other major problems facing humanity" first... I am going to criticize this attitude as radically out of touch with reality in multiple ways. If it helps, I had similar thoughts once. I didn't want to do the "other major problems" first, but I wanted to do them at the same time.

So, let's see what's wrong with this picture. First: what makes you think that you get a say on any of this - either the course of world events, or the rate at which the human race tries to overcome its biological limitations? Are you in a position, of influence, wealth, knowledge, ability, to make a difference to anything much outside your own life? Or are you just an atom of will slowly realizing that it was born into a universe completely at odds with the sort of reality it would like to be inhabiting, an atom of will which is able to express this preference verbally, but which is not able to act on it in any way?

That is the harshest thing that needs saying. Most people who start young with big dreams more or less completely abandon them, because it's hard to make them actually happen. A few people persist at the price of everything else, perhaps without even realizing that this is what they are doing, or understanding why life seems to be so difficult; I'd place my younger self in this category. It would be an extremely rare and fortunate person who neither compromises their ideals nor suffers because of this. The world around you is waiting for you to abandon any idea of immortality or a better world, in order to focus on more achievable things like making money and having a relationship, and it is set up to support this sort of activity, not the other sort.

Next, let's consider what's implied by solving those "other major problems". By the sound of it, here you mostly mean "unsustainability" - overpopulation, maybe resources running out - though you could also have in mind poverty, war, and who knows what else.

Just to be realistic, let's remember that these problems have been around for the whole of human history. Saying that before you get rid of X (in this case, death), you want to get rid of Y first, where Y is at least half a dozen things that have existed forever despite the intention of numerous of your predecessors to get rid of them - that sounds like a formula for never getting around to dealing with X.

Also, the human race does not inhabit one single rational polity run by philosopher-kings. It consists of about 200 sovereign states whose governments are already always in crisis management mode, because they're dealing with insurgencies, culture wars, the exigencies of the global market, and so on. Here's a tip: do you want to know how fanatical movements get created, the sort that wage wars, demonize opponents, and so on? One way they get created is that someone decides to solve all the world's problems at once, and that becomes such an overwhelmingly positive outcome that people become desperately willing to do anything in order to make it possible. The world is so far beyond the control of any individual atom of will, or even any political coalition of them, that those who insist on having an impact at the highest levels of the historical process generally have to adopt the most extreme available measures.

Another problem is that technological development is not going to wait for you to try to achieve a perhaps impossible planetary "equilibrium" in which distressing things are no longer happening in any corner of the world. The flip side of this is that peak oil, new diseases, and terrorists with WMDs are not going to stop a singularity from happening. Despite what the myths of planetary Internet mind-union say, the world does not stand together or fall together, and it especially does not advance together. The world could be having a major eco-economic crisis, but any sufficiently advanced country can at any time pull a North Korea, put up a wall, and retire from the globalization game without sacrificing its technological culture. Two-thirds of the world could go Mad Max, and the progress of the final third would still be far more than enough to drag us all into the AI/nanotech zone. So if you do choose to maintain an interest in ultra-high-tech post-human possibilities, get used to doing this even while the slums on the other side of town are on fire, metaphorically speaking. You can always choose to be a firefighter instead, but that really does mean you will no longer be very important in solving those more futuristic problems.

I had some other things to say but this is already more than enough.

I think just about everything in the above comment is predicated on a grossly incorrect reading of the original post. So far as I can tell:

Raemon was using "getting rid of death" not to mean "magically ensuring that no one ever dies at all, even on account of the end / heat-death of the universe" but "eliminating, or very nearly eliminating, death through old age and disease", and nothing s/he said requires the former meaning.

Raemon was not claiming to have any substantial ability to influence whether or not the human race abolishes death, or when, nor to play a big role in solving all those other problems. (And: discussing whether something would be a good or a bad thing doesn't in the least imply any claim to be able to influence it.)

Raemon was not claiming to have a politically feasible way of solving any of those problems. In particular, s/he was not claiming that the world is a single rational polity run by philosopher-kings, s/he was not proposing to form a fanatical extremist movement, etc., etc., etc. (And: discussing whether something would be a good or a bad thing doesn't in the least imply any claim that there's an easy way to bring it about.)

Thank you. Both for understanding what I was talking about (more accurately, what I wasn't), and for making the effort to use gender neutral wording. (As it happens I am male, but it's something I care about which happens to be on my mind right now).

I totally understand why death is a big deal. I'm not saying we should wait to fix it. I understand emphasizing that because it's not something that generally gets press. But there's a lot of other issues out there that need fixing as well, and not everyone is the sort of person who has the skillset/interest in fixing the death issue directly. My target audience was internet bloggers, whom I thought were disproportionately talking about one issue. Not government officials.

Mostly, though, I honestly just wanted to know what the counterarguments to overpopulation were, and they have been explained to my satisfaction.

I don't know how much this reflects the site consensus, but I have two responses to the overpopulation argument.

Coarse response: What's the problem with overpopulation? It's that resources get used up and the world can't support everybody. And what's the problem with that? People die. Eliminating death by aging with the result that some starve may still result in less total death.

Sophisticated response: As education improves and wealth increases, fertility rates fall. Much of the first world has less than replacement birth rate; US population would be shrinking were it not for immigration. World population is expected to peak around 9 billion around 2050 (source: The Economist Magazine). So as standard of living increases, population growth may slow to the point that it won't be too hard to legislate limits on reproduction.

[-][anonymous]10y 2

Sophisticated response: As education improves and wealth increases, fertility rates fall. Much of the first world has less than replacement birth rate; US population would be shrinking were it not for immigration. World population is expected to peak around 9 billion around 2050 (source: The Economist Magazine). So as standard of living increases, population growth may slow to the point that it won't be too hard to legislate limits on reproduction.

This is temporary. Genetically speaking we are experiencing a challenge about as severe as that of a major climate shift. Natural selection will eventually cause population growth to continue if left uninterrupted.

The following is by my estimation being selected for:

  • Later menopause
  • Earlier menarche and sexual maturation coupled with stronger sexual drive early in life
  • High time preference
  • Lower IQ (with stronger selective pressures being experienced by women unfortunately)
  • Besides lower IQ there is Increased neurodiversity, especially the space that makes academic success harder but does not adversely affect socialization in general.
  • Stronger ability to rationalize "failure" and convincingly communicate such rationalizations (the price of self-delusion has fallen relative to the ancestral environment)
  • Desire for children
  • Religiosity

Almost any categorization of the population will find groups that excel according to these criteria outgrowing others. Even memetic transmission with its low reliability has produced impressive advantages for groups like say the Amish.

What's the problem with overpopulation? It's that resources get used up and the world can't support everybody. And what's the problem with that? People die.

No, my problem with overpopulation is that people suffer. (And then die in addition to that).

That said, I understand and mostly share your optimism about the results of increased standard of living.

No, my problem with overpopulation is that people suffer. (And then die in addition to that).

That position would only make consistent sense in a world where aging doesn't cause suffering. Have you been to a nursing home recently? Seen how many painkillers senescent individuals tend to take on a regular basis to mask the symptoms of their diseases?

But if painkillers are a satisfactory solution for the suffering of aging, why do we not propose them to address the suffering of starvation? Surely dying of hunger is not such a bad subjective experience if you can do it while high on morphine.

If we have a way to prevent everyone in the world from suffering due to malnutrition, I'd consider that a win either way.

Overpopulation is last millennium's meme. The birth rate in most regions of the world has now crashed below extinction level, and the remainder look likely to follow as soon as they develop their economies. Frankly, apart from the compelling moral and humanitarian reasons for opposing death, from the most dispassionate 'good of the species as a whole' perspective, a major reason we need to do so is as a stopgap against the ongoing population decline.

The birth rate in most regions of the world has now crashed below extinction level

Less dramatic version: The fertility rates in Europe, North America, Oceania and the Far East have gradually fallen to or slightly below replacement level, and would lead to "extinction" in the UK in a mere 5000 years assuming everything stays the same. (Extinction in the USA would follow in a mere infinity years.)

World fertility rate is still about 2.5 children per woman.

and the remainder look likely to follow as soon as they develop their economies.

Which so far just means competing for the same depleted resources without making noticeable progress towards developing new ones. 'Overpopulation' is relative to the resource base. Ours may be shrinking alarmingly over the coming years.

ongoing population decline.

Less dramatic version: the ongoing decline in the population growth rate. In the UN's 2004 projection, even in the "low" scenario, the population peak isn't reached until about 2040. The "medium" scenario puts it at 2080 (which is surely beyond the point at which current predictions mean anything.)

Less dramatic version: The fertility rates in Europe, North America, Oceania and the Far East have gradually fallen to or slightly below replacement level, and would lead to "extinction" in the UK in a mere 5000 years assuming everything stays the same. (Extinction in the USA would follow in a mere infinity years.)

Where are you getting those figures? Here are some from http://en.wikipedia.org/wiki/List_of_sovereign_states_and_dependent_territories_by_fertility_rate

  • United States: 2.05 - already less than replacement level, and even this figure as I understand it depends on high birthrates among early generation immigrants, and would rapidly drop if not continually fed by immigration.

  • United Kingdom: 1.82

  • Russia 1.34

  • Japan: 1.27

A birth rate as low as Russia or Japan means the population drops by more than half in two generations. That means even if the status quo were maintained all the way to the end - which of course it can't be - the sum and total of all future lives lived after that point, the sum of all joy and laughter, of all thought and effort and progress driven by that thought and effort, will be less than in these next two generations.

Two generations. That long ago, we already had nuclear energy, satellites, transistors, jet airliners and mainframe computers. Two generations just isn't very much time in which to get things done.

Which so far just means competing for the same depleted resources without making noticeable progress towards developing new ones.

On the bright side, I have noticed some progress - the graphs for solar energy in recent decades are looking encouraging, for example. More importantly, we are still seeing progress in critical foundation areas like robotics, biotech and nanotech; those are the things we need to master if we want to take off before our runway runs out.

'Overpopulation' is relative to the resource base. Ours may be shrinking alarmingly over the coming years.

It may.

One of the things I'm worried about is this:

I would like to think if we don't make it this time around, civilization can pick itself back up for another shot. Even difficult goals can often be attained if you get to try as many times as you want.

But the development of industrial technology was based on abundant and easily obtainable deposits of coal, oil etc. We haven't run out of those yet - but the easily obtainable deposits thereof are mostly already gone. Yes, we can imagine a future reconstruction of industrial civilization by other pathways, but would that really happen?

Nobody knows. I'd rather not try the experiment. I think we need to behave as though we have to take off on this attempt.

[-][anonymous]10y 1

I would like to think if we don't make it this time around, civilization can pick itself back up for another shot. Even difficult goals can often be attained if you get to try as many times as you want.

But the development of industrial technology was based on abundant and easily obtainable deposits of coal, oil etc. We haven't run out of those yet - but the easily obtainable deposits thereof are mostly already gone. Yes, we can imagine a future reconstruction of industrial civilization by other pathways, but would that really happen?

Nobody knows. I'd rather not try the experiment. I think we need to behave as though we have to take off on this attempt.

I worried about this for a long time, but recently I've been thinking that perhaps this isn't such a high barrier to jump.

If a form of civilization based on agriculture is maintained after the technological fallback the next time around it seems plausible that we may also have a higher IQ and be generally better adapted to life in mass society. Human brains are pretty good at finding substitute resources.

Of course Hanson would point out that the farmer forager value compromise would skew more heavily towards farmers the next time. But since I seem to have a slightly more farmer predisposition than most and counting on the seemingly reasonable presupposition that wealth will again make them more forager than might seem at first glance possible, I would be pretty ok with this scenario.

Except for its effect on the odds of my revival.

I don't think there's a compelling reason to KEEP the population at its current levels, until/unless everyone's basic needs are being met.

The kind of specialized investment required to maintain and develop technology at our current level requires market sizes not much smaller than the current ones (e.g. check out what a modern microchip factory costs.) If we didn't mind going back to town-blacksmith tech levels, sure, population could get a lot smaller - but that would, among other problems, kill off the idea of meeting everyone's basic needs.

I think there's a pretty large range between blacksmith levels and modern society. I'm pretty sure I could be happy living with 50s era technology. I don't know what strides have been made in medicine and agriculture since than that might be contingent on having 6 billion people instead of 3.

On the other hand, 50s era technology depended heavily on nonrenewable resources. (Not just the obvious fossil fuels either, e.g. I've seen it claimed, can't verify or refute of my own knowledge, that agriculture is going to start running into trouble after another few decades of depletion of phosphate deposits unless we find a solution.)

We have good reason to believe resource depletion problems can be solved if we continue to make progress rapidly enough. But how long a window of time we have, and how long things could be kept ticking along at industrial era tech level, nobody knows, and I'd rather not find out the hard way.

rwallace:

The kind of specialized investment required to maintain and develop technology at our current level requires market sizes not much smaller than the current ones (e.g. check out what a modern microchip factory costs.)

I've read this claim many times, but I've never seen a convincing argument for what exactly would count as "much smaller." My impression is that three orders of magnitude smaller would almost certainly be too small, two orders of magnitude possibly also, but I'm not at all sure about one order of magnitude.

Is there some kind of critical analysis of this I can read? What technologies are the most vulnerable to stagnation at lower scales? It seems like a lot of technological progress is one-way, in that markets have already selected for the most efficient form of technology. It could then be scaled down significantly. Computer chips maybe not so much, but what about (say) farming?

I haven't come across any reasonably complete and in-depth analysis of this (just a bunch of scattered and more or less tangential fragments to which I didn't keep a list of references) - if anyone else has links to such, I'd be interested in taking a look.

One big problem with regarding technological progress as one-way is that a great deal of technology relies on nonrenewable resources. To take your example of farming, today's agricultural techniques rely very heavily on fossil fuels and perhaps on less obvious resources: consider phosphate deposits, soil erosion and pesticide resistance - all problems which might be solved easily with sufficiently advanced technology, but not necessarily otherwise.

The conclusion is that it's hard to know how much time we have, but one thing is clear: stagnation is not stasis. It is, in the end, death.

[-][anonymous]10y 0

Death is an enemy that should be vanquished, but it's not an enemy I'm prepared to destroy until I'm satisfied that we'll be able to handle the consequences.

Then you think being immortal would be worse than being mortal until we solve these other problems, such as overpopulation?

Why? At the very least, any immortal who prefers dying to living in an overpopulated world will always have the option of suicide.

A thought-puzzle I find amusing: in a world with infinite resources (e.g. an infinite cellular automaton, such as the one in a certain Greg Egan book), there exists a civilization-wide reproduction strategy such that (1) the population will grow only polynomially in time, and (2) every agent will have infinitely many children.

It's actually rather simple, and doesn't require slowdowns of subjective time. I thought I'd leave it to you as a quick exercise...